This study is aimed at developing a predictive model of pervaporation performance, particularly permeation flux, as a function of physicochemical properties of permeant-membrane material and process conditions. A new approach was introduced that the developed model took into account all the steps of transport mechanism in pervaporation, i.e., sorption, diffusion, and desorption. A new term so-called permeation conductance was used and was correlated to selected affecting physicochemical properties and process conditions on the above three steps through dimensional analysis. The permeation flux (J(i)) was modeled as a multiplication of an overall permeation conductance (k(overall,i)) and a driving force of feed concentration (C(F,i)). A large data based was assembled from the literature and our experiments; 87% of it was used for model training, and the remaining was for testing. The model was found to give fairly good predictions on pervaporation performance. The relative ratio of molecular interactions (solubility parameter) of permeant-permeant over that of permeant-membrane (Delta Sol(ij)/Delta Sol(im)) and the operating temperature (T) were found to be the most influencing physicochemical property and process condition on the permeation flux. (C) 2011 Elsevier B.V. All rights reserved.